There is limited information published regarding treatment of disfluencies in Autism Spectrum Disorders (ASDs). However, both typical (Paul et al., 2005; Shriberg et al., 2001; Sisskin, 2006) and atypical (Hietala & Spillers, 2005; Scott, Grossman, Abendroth, Tetnowski & Damico 2006; Sisskin, 2006; Stribling, Rae & Dickerson, 2007) patterns of stuttering have been noted in some individuals diagnosed with ASDs. Although at this time we can only speculate about underlying causes of disfluencies, there is supporting data within other areas of communication disorders and from research in ASDs that can guide our decision making for effective treatment.

This paper is meant to bridge the gap between research findings and intervention strategies for best practice in treating these unusual cases. Our goal is to provide the clinician with a model of critical thinking based on the data currently available. The following four cases represent the various ASD diagnoses, levels of functioning, and types of disfluencies with which a clinician might be presented. They are compiled from cases on which the authors consulted. The reader is instructed to bear in mind that although these cases are presented to illustrate the range of disfluency symptoms that a clinician may encounter, the patterns that follow are not exclusive to each category of ASD in which they are presented.

Case Descriptions

Case One: Autistic Disorder

Background:
C1 was an 8-year-old male with a diagnosis of autism who attended a special education program in a public school. He received speech and language services at school and attended a university speech and language clinic for additional services. He demonstrated average nonverbal cognitive functioning based on the results of the Leiter-R International Performance Scale. Criterion referenced tests and inventories revealed severe delays in personal/social, adaptive, motor, communication and cognitive domains. Academic readiness skills were an area of relative strength, and appeared to be between one and two years delayed. C1 demonstrated poor nonverbal communication skills, for example, he did not exhibit eye-gaze shifts during communicative acts. Parents reported stereotypical behavior, although none were observed during the observation period. Onset of speech disfluencies were reported by parents and observed by clinicians during the same time period. Onset appeared to be sudden, at the age of 8. There was no reported history of stuttering or other speech or language disorders in the family.

Profile of Disfluency:
C1's speech was sampled during a speech and language therapy session while engaging in a variety of clinician and child-directed activities. Speech contexts included spontaneous speech during play and responding to questions during a book sharing activity. C1 demonstrated stuttering-like disfluencies characterized by initial sound repetitions, "t-t-t-t-t-turn around". Stuttering was judged to be moderate in terms of frequency; however, disfluencies were often characterized by up to 15-20 iterations (repetition of the same sound, syllable or word), which might have added to a perception of greater severity. Physical concomitant behaviors during moments of stuttering were mild and included lowering his head, phonating on residual air, and jerking his torso when releasing the disfluency to complete a word. There appeared to be tension during phonation and respiration while disfluent. There did not appear to be any awareness of disfluency or difficulty speaking, which may explain his persistence in working through a repetition of up to 20 iterations. Disfluencies occurred in utterances of all lengths (one-, two- and three-word utterances), but most often occurred during spontaneous, non-imitated speech.

Language Profile:
Criterion reference testing revealed a receptive language age of 44 months and an expressive language age of 32 months. C1's speech sample consisted of both imitated and spontaneous, self-generated utterances up to three words in length. Imitated utterances included immediate echolalia, scripted phrases (perhaps delayed echolalia) and repetition of the clinician's modeled target utterance during an instructional activity (often prompted). Phrases that appeared to be scripted, for example, "Hi everybody" and "Turn around", were uttered multiple times during the sample. C1 used picture communication symbols to aid in both expression and comprehension. He was able to follow simple directions and answer questions about items in his immediate surroundings. During the previous few months, teachers had reported an increase in spontaneous communication (asking to go to the bathroom), more frequent responses to bids for communication from others, and greater interaction with peers (extended play schemes with other children). While peer interactions were not always socially appropriate, C1 had a preferred friend at school.

Discussion:
C1's disfluency pattern was typical of early stuttering. Symptoms generally appear around the age of 3 (mean), with the majority of cases reporting rapid or intermediate onset as opposed to gradual onset (Yairi & Ambrose, 2005). The type of disfluency demonstrated by C1 (repetitions) was also typical of developmental stuttering at early stages, as 50% of the cases studied by Yairi and Ambrose (2005) began with repetition, and most reported accompanying force or tension and at least one physical concomitant behavior. C1's onset of stuttering was at age 8; however, this was the time when his language skills were estimated to be at the 3-year-old level, generally a time of language growth. If it is true that linguistic demands affect childhood fluency (Hall, Wagovich, Bernstein Ratner, 2007), and there is a trade-off of resources (language demands and motoric fluency), then we might speculate that C1's disfluency would follow a typical path of developmental stuttering.

Treatment Considerations:
Treatment decisions for C1 might include evaluating the factors considered important for decision making for typically developing children in the early months of stuttering. Since the majority of children who begin to stutter during the preschool years recover without intervention, it would therefore be important to examine the risk factors for persistence and recovery (Yairi & Ambrose, 2005) as they relate to C1. Positive factors included time since onset (C1 was within the early window of expected recovery) and negative family history for stuttering. Current research does not address how a severe communication disorder and limited perspective-taking skills characteristic of autism would impact recovery from stuttering or the progression of the disorder. While there is no evidence that children who recover from stuttering have stronger language skills than those who persist (Yairi and Ambrose, 2005), the persistent group may demonstrate more variability in their expressive language profiles than those who recovered (Watkins & Yairi, 1997). More questions arise if a decision is made to begin treatment. Studies have shown that a considerable percentage of children who stutter have a concomitant language disorder (see Hall et al., 2007 for a review), but we do not know the best course of treatment for children with autism or the factors that would affect their ability to respond to and benefit from current treatment approaches.

The decision was made to delay treatment for stuttering for C1 because treatment priorities continued to emphasize improved functional communication and social language. This was a fairly easy decision based on the recent onset of symptoms and the lack of impact for C1 in cognitive and emotional domains. However, it would be interesting to speculate about the various treatment approaches that would produce positive outcomes had the decision been made to treat stuttering. Bellon-Harn, Harn & Watson (2007) reported on an approach that utilized rule learning for targeting prosody in a case study of an 8-year-old boy with High Functioning Autism. Rules for targeting excessive pausing, for example, included "not using big pauses between words". The strategy of using rules for good speech was deemed to be a good match for this child who adhered strictly to rule-systems in academic and community settings. Rule learning, combined with modeling and feedback was successful in reducing the prosodic features of silent and audible pausing as well as syllable lengthening for this child. This might be a viable approach to take for C1 for reducing repetitions associated with stuttering.

Case Two: Autistic Disorder with Intellectual Disability

Background:
C2 was a 20-year-old male with a diagnosis of Autistic Disorder. While he was unable to conform to standardized tests of intellectual functioning, school psychological testing placed him in the moderate range of mental retardation. He communicated verbally with one- and two-word utterances, mainly for the purpose of behavior regulation, specifically, requesting items or actions. C2 presented with a variety of stereotypical behaviors including rocking and hand movements. He engaged in rituals that appeared to be compulsive in nature, including touching windows, adjusting the orientation of videodisks, and rotating his body prior to sitting. He attended a school program that emphasizes functional academics and prevocational skills, and received speech and language therapy, occupational therapy, and behavior consultation services as part of his school program. Onset of disfluency was sudden, beginning at age 18. Family history was positive for stuttering.

Profile of Disfluency:
C2's speech was sampled in a variety of daily activities in his home setting including leisure activities and while eating a meal. He displayed disfluencies on approximately 50% of intended words in the speech sample. Half of the disfluencies were whole-word repetitions, "cheese-cheese", and half were sound/syllable repetitions, "bu-bugs." Utterances often contained a combination of whole-word repetition and part-word repetition, "juice-juice, ju-ju-ju-ju-juice-juice", and on occasion, strings of disfluencies were mitigated with a variation of the sound pattern as in the following example, when C2 requested to watch TV, " Teevee-teevee-teebee-bee-bee-teevee, teevee".

Approximately 90% of disfluent utterances occurred on highly motivated requests. A greater number of iterations (for both whole- and part-word repetition) were noted when anxiety and request intensity were high. Disfluency occurred with less frequency on utterances for the purpose of joint attention (answering questions) and social interaction (greeting). No physical concomitant behaviors were present during disfluency, and no awareness of disfluency was apparent.

Language Profile:
C2's communication was limited in terms of both functions and means. As mentioned above, he communicated with one- and two-word utterances. Longer utterances were present in his language sample but were more likely representative of a gestalt mode of language acquisition in which multi-word utterances function as a single word or as a label for an event (Wetherby & Prizant, 2000). In this case, it is likely that C2 began using the utterances as a verbal repetition strategy (echolalia) and they continued to function as a single grammatical unit. For example, "go out in the car"functioned as a consistently used request to go for a ride. Vocabulary knowledge was a relative strength for C2. He understood a wide variety of nouns and verbs, and followed simple directions in familiar contexts. He benefited from use of picture communication symbols to aid in comprehension in daily activities and during instructional activities, but did not spontaneously use picture symbols to aid in verbal expression, even when they were provided.

Discussion:
While C2's disfluency types (part- and whole-word repetition) were those associated with stuttering-like disfluencies, the quality of the repetition was not typical of developmental stuttering. C2 continued to repeat word units with what appeared to be a different intent than seen in developmental stuttering. Furthermore, the number of iterations of the unit appeared to be related to the delay in compliance of the conversational partner in granting the verbal requests. In reviewing the literature, it appears that C2's disfluency profile directs us to three potential explanations, presented here for speculation.

The first explanation is that C2's disfluency was a form of palilalia. Palilalia is associated with a variety of neurogenic disorders and is described as involuntary, spontaneous repetitions of the same word or phrase. The units of repetition are longer and the number of iterations is more numerous than those seen in developmental stuttering (Van Borsel, Bontinck, Coryn, Paemeleire & Vandemaele, 2007). The repetitions described in the case of a 60-year-old man diagnosed with idiopathic Parkinson disease consisted of "several trains of repetition with short pauses between them" (Van Borsel et al., 2007), very much like the strings of repetitions evidenced in C2's speech sample. Buckingham and Christman (2004) describe examples of recurrent perseveration in aphasia in which phonemic carryover results in newly formed hybrid words. Their example of rime perseveration appears to match some of C2's phonemic alterations in multiple repetitions.

Another explanation was that the disfluencies in C2's speech served as a verbal repetition strategy, similar to echolalia, with a particular function. In this case, C2 repeated the single-word request until it was granted. The number of repetitions increased, along with anxiety, during the period between the initiation of the verbal request and the granting of the request by the communication partner. The repetitions stopped once the function of the communication was complete. This pattern is very similar to the forms of repetition exhibited by a 16-year-old girl with autism in a case study by Stribling, Rae, & Dickerson (2007). The young woman, Helen, demonstrated two forms of repetition described as prior-turn repeats (repetition of another's prior talk) and within-turn repeats (repetition of one's own prior talk). The authors attribute Helen's within-turn repeats to a form of palilalia. However, as palilalia is considered to be involuntary, C2's within-turn repeats may be more likely a form of immediate echolalia. Echolalia may be associated with a communicative or non-communicative function. For example the echo may be a placeholder for a conversational turn (Prizant & Rydell, 1984). When asked, "Do you want lunch?" an echolalic response, "Do you want lunch" or a form of the utterance, "want lunch" (an example from C2's sample) serves the function of an affirmation rather than "yes", which was not a part of C2's expressive language repertoire.

A final explanation was drawn from the literature investigating repetition in typically developing speech. In her classic analysis of a toddler's pre-sleep monologues, Weir (1962) described babbling as an example of language practice or articulatory exercise. Other researchers have described the infant monologues of Emily, the case studied in Narratives from the Crib, as having a problem solving, or self-regulating function (Nelson, 1989). A study of the private speech of young children in their homes revealed that 89% of self-talk was self-regulatory in nature, to enhance recall or help in choice making. Private speech also increased with task difficulty (Patrick & Abravanel, 2000). It is worth considering that C2's disfluency may have been a form of practice, or perhaps self-regulation, particularly because monologue and forms of self-talk have been interpreted as having a self-regulatory function (Prizant, 1984) for some individuals with autism.

Treatment Considerations:
While C2's speech disfluencies were observed by his parents, no mention of these symptoms were made by his school teachers or therapists despite the frequency of disfluency in verbal communication. This is not unusual considering that the identified priorities for treatment included establishing a consistent functional verbal communication system, expanding the repertoire of functions in communication, and development of nonverbal communication for social interaction. Direct targeting of disfluency was not a priority at the time.

It would be interesting to consider possible treatment directions had functional communication not been the clear priority. If we hypothesized that C2's disfluencies represented a verbal repetition strategy with communicative intent, similar to echolalia, then increased disfluencies in C2's speech might be considered an indication of language growth. In this case disfluency could be seen as a "placeholder" with the pragmatic intent of "persisting". For example, repeating the command, "Bye-b-b-bye-bye-bye-bye-!Šbye-bye!" might have been replaced with, "OK, go already!" by a young adult without autism. It is important to note that if C2's repetitions were determined to be serving a communicative function, it would not be appropriate to target elimination of the repetition at the cost of interfering with pragmatic language development. If however, disfluencies were interpreted as self-talk or motor practice, as described earlier, we might expect C2 to move through this stage, as young children do as skills improve. Finally, if disfluency is a form of palilalia or functions similar to a motor stereotypy, we might see improvement with medication aimed at reduction of stereotypical behavior, for example, Risperidone, an atypical antipsychotic drug prescribed to reduce irritability and stereotypical behavior in children with autism (Chavez, Chavez-Brown, & Rey, 2006). This would be a medical decision that would have to be weighed against possible consequences of side effects or behavior changes that often result from these drugs.

Note: Two additional cases of ASD and disfluency are described in Part II of this article: Speech Disfluency in Autism Spectrum Disorders: Clinical Problem Solving for Pervasive Developmental Disorder, Not Otherwise Specified and Asperger Syndrome.

References

Bellon-Harn, M.L., Harn, W.E. & Watson, G.D. (2007). Targeting prosody in an eight-year-old child with high-functioning autism during an interactive approach to therapy. Child Language Teaching and Therapy, 23, 157-79.

Buckingham, H.W., Christman, S.S. (2004). Phonemic carryover perseveration: Word Blends. Seminars in Speech and Language, 25, 363-73.

Chavez, B., Chavez-Brown, M. & Rey, J.A. (2006). Role of risperidone in children with autism spectrum disorder. Annals of Pharmacology, 40, 909-16.

Hall, N.E., Wagowich, S.A., Bernstein Ratner, N. (2007). Language considerations in childhood stuttering. In Conture, E.G. & Curlee, R.F., (Eds.). Stuttering and Related Disorders of Fluency, 3rd edition. New York: Thieme.

Heitala, A. and Spillers, C. (2005). Disfluency patterns in children with autism spectrum disorders. Poster presentation, Annual Convention of the American Speech-Language Hearing Association, San Diego.

Nelson, K. (Ed.) (1989). Narratives from the Crib. Cambridge: Harvard University Press.

Patrick, E. & Abravanel, E. (2000). The self-regulatory nature of preschool children's private speech in a naturalistic setting. Applied Psycholinguistics, 20, 45-61.

Paul, R., Shriberg, L.D., McSweeny, J., Cicchetti, D., Klin, A., & Volkmar, F. (2005). Brief report: Relations between prosody performance and communication and socialization ratings in high functioning speakers with autism spectrum disorders. Journal of Autism and Developmental Disorders, 35, 861-69.

Prizant, B.M. & Rydell, P.J. (1984). Analysis of functions of delayed echolalia in autistic children. Journal of Speech and Hearing Research, 27, 183-92.

Scott, K. S., Grossman, H. L., Abendroth, K. J., Tetnowski, J. A., & Damico, J. S. (2006). Asperger syndrome and attention deficit disorder: Clinical disfluency analysis. In J. Au-Yeung & M.M. Leahy (Eds.). Research, treatment, and self-help in fluency disorders: New Horizons. Proceedings of the Fifth World Congress on Fluency Disorders. Dublin, Ireland: International Fluency Association.

Sisskin, V. (2006). Speech disfluency in Asperger Syndrome: Two cases of interest. Perspectives on Fluency and Fluency Disorders, 16, 12-14.

Shriberg, L.D., Paul, R., McSweeny, J.L., Klin, A, Cohen, D.J., & Volkmar, F.R. (2001). Speech and prosody characteristics of adolescents and adults with high- functioning autism and asperger syndrome. Journal of Speech, Language, and Hearing Research, 44, 1097-1115.

Stribling, P., Rae, J., & Dickerson, P. (2007). Two forms of spoken repetition in a girl with autism. International Journal of Language and Communication Disorders, 42, 427-44.

Van Borsel, J., Bontinck, C., Coryn, M., Paemeleire, F., & Vandemaele, P. (2007). Acoustic features of palilalia. Brain and Language, 101, 90-96.

Watkins, R.V. & Yairi, E. (1997). Language production abilities of children whose stuttering persisted or recovered. Journal of Speech, Language, and Hearing Research, 42, 1125-35.

Weir, R.H. (1962). Language in the Crib The Hague: Mouton & Co.

Wetherby, A.M. & Prizant, B.M. (2000). Autism Spectrum Disorders: A Transactional Development Perspective. Baltimore: Paul H. Brooks.

Yairi, E. & Ambrose, N.G. (2005). Early Childhood Stuttering: For Clinicians by Clinicians. Austin, TX: ProEd.

You can post Questions/comments about the above paper to the author before October 22, 2007.